In the code, the analysis should be done using python, numpy and/or pandas methods rather than hardcoding results manually.
Here are summary statistics of the traffic signs data set:
The size of training set is 34799
The size of the validation set is 4410
The size of test set is 12630
The shape of a traffic sign image is (32, 32, 3)
The number of unique classes/labels in the data set is 43
As follow, the each datasets have different distributions of frequency.
In the traing dataset, there is near ten times difference among 43 labels(classes). Therefore it is possible that equal training can not be done with this dataset.
Moreover, some labels may have shortage of sample number to adequately train.
各ラベルの分散の違い
平均値の分布 画素値分散の分布
2. an exploratory visualization of the dataset.
Here is an exploratory visualization of the data set.
Dataset seems to be augmented via some image processing like changing brightness, contrast, chroma and cropping position.
It is a bar chart showing how the data ...
各クラスの平均画像
各クラスの平均画像を作る
最初の正規化検討
前処理の説明と、その理由
各クラスの平均画像の変化
間違えられたクラスの頻度を見る
間違えられたクラスの頻度を見る
ダメな画像を見付ける
After the 1st trial of LeNet-5 with above normalization
1. Describe how you preprocessed the image data. What techniques were chosen and why did you choose these techniques? Consider including images showing the output of each preprocessing technique. Pre-processing refers to techniques such as converting to grayscale, normalization, etc. (OPTIONAL: As described in the "Stand Out Suggestions" part of the rubric, if you generated additional data for training, describe why you decided to generate additional data, how you generated the data, and provide example images of the additional data. Then describe the characteristics of the augmented training set like number of images in the set, number of images for each class, etc.)
Accuracy on the validation set is 0.93 or greater.
As a first step, I decided to convert the images to grayscale because ...
Here is an example of a traffic sign image before and after grayscaling.
alt text
As a last step, I normalized the image data because ...
I decided to generate additional data because ...
To add more data to the the data set, I used the following techniques because ...
Here is an example of an original image and an augmented image:
alt text
The difference between the original data set and the augmented data set is the following ...
2. Describe what your final model architecture looks like including model type, layers, layer sizes, connectivity, etc.) Consider including a diagram and/or table describing the final model.
My final model consisted of the following layers:
Layer
Description
Input
32x32x3 RGB image
Convolution 3x3
1x1 stride, same padding, outputs 32x32x64
RELU
Max pooling
2x2 stride, outputs 16x16x64
Convolution 3x3
etc.
Fully connected
etc.
Softmax
etc.
3. Describe how you trained your model. The discussion can include the type of optimizer, the batch size, number of epochs and any hyperparameters such as learning rate.
To train the model, I used an ....
4. Describe the approach taken for finding a solution and getting the validation set accuracy to be at least 0.93. Include in the discussion the results on the training, validation and test sets and where in the code these were calculated. Your approach may have been an iterative process, in which case, outline the steps you took to get to the final solution and why you chose those steps. Perhaps your solution involved an already well known implementation or architecture. In this case, discuss why you think the architecture is suitable for the current problem.
My final model results were: * training set accuracy of ? * validation set accuracy of ? * test set accuracy of ?
If an iterative approach was chosen: * What was the first architecture that was tried and why was it chosen? * What were some problems with the initial architecture? * How was the architecture adjusted and why was it adjusted? Typical adjustments could include choosing a different model architecture, adding or taking away layers (pooling, dropout, convolution, etc), using an activation function or changing the activation function. One common justification for adjusting an architecture would be due to overfitting or underfitting. A high accuracy on the training set but low accuracy on the validation set indicates over fitting; a low accuracy on both sets indicates under fitting. * Which parameters were tuned? How were they adjusted and why? * What are some of the important design choices and why were they chosen? For example, why might a convolution layer work well with this problem? How might a dropout layer help with creating a successful model?
If a well known architecture was chosen: * What architecture was chosen? * Why did you believe it would be relevant to the traffic sign application? * How does the final model's accuracy on the training, validation and test set provide evidence that the model is working well?
Test a Model on New Images
Test a Model on New Images
Acquiring New Images: German Traffic signsから5つの画像を加え、図示し、分類が難しいかどうか論じる
Performance on New Images: キャプチャ画像をテストすつ時のモデルの性能
Performance on New Images: テストセットに対する、新しい画像に対する性能
Model Certainty - Softmax Probabilities:
1. Choose five German traffic signs found on the web and provide them in the report. For each image, discuss what quality or qualities might be difficult to classify.
Here are five German traffic signs that I found on the web:
The first image might be difficult to classify because ...
2. Discuss the model's predictions on these new traffic signs and compare the results to predicting on the test set. At a minimum, discuss what the predictions were, the accuracy on these new predictions, and compare the accuracy to the accuracy on the test set (OPTIONAL: Discuss the results in more detail as described in the "Stand Out Suggestions" part of the rubric).
Here are the results of the prediction:
Image
Prediction
Stop Sign
Stop sign
U-turn
U-turn
Yield
Yield
100 km/h
Bumpy Road
Slippery Road
Slippery Road
The model was able to correctly guess 4 of the 5 traffic signs, which gives an accuracy of 80%. This compares favorably to the accuracy on the test set of ...
新たな画像5枚を選んで教師画像に加える
3. Describe how certain the model is when predicting on each of the five new images by looking at the softmax probabilities for each prediction. Provide the top 5 softmax probabilities for each image along with the sign type of each probability. (OPTIONAL: as described in the "Stand Out Suggestions" part of the rubric, visualizations can also be provided such as bar charts)
The code for making predictions on my final model is located in the 11th cell of the Ipython notebook.
For the first image, the model is relatively sure that this is a stop sign (probability of 0.6), and the image does contain a stop sign. The top five soft max probabilities were
Probability
Prediction
.60
Stop sign
.20
U-turn
.05
Yield
.04
Bumpy Road
.01
Slippery Road
For the second image ...
(Optional) Visualizing the Neural Network (See Step 4 of the Ipython notebook for more details)
1. Discuss the visual output of your trained network's feature maps. What characteristics did the neural network use to make classifications?
TODO
[X] Load the data set (see below for links to the project data set)
[ ] Explore, summarize and visualize the data set
[ ] Design, train and test a model architecture
[ ] Preprocessing: preprocessing techniques used
[ ] Preprocessing: and why these techniques were chosen.
[ ] Model Architecture: the type of model used, the number of layers, the size of each layer.
[ ] Model Architecture: Visualizations emphasizing particular qualities of the architecture
[ ] Model Training: how the model was trained by discussing, what optimizer was used/batch size/number of epochs/values for hyperparameters.
[ ] Solution Approach: the approach to finding a solution.
[ ] Solution Approach: Accuracy on the validation set is 0.93 or greater.
[ ] Acquiring New Images: five new German Traffic signs found on the web, and the images are visualized.
[ ] Acquiring New Images: Discussion is made as to particular qualities of the images or traffic signs in the images that are of interest,
[ ] Acquiring New Images: such as whether they would be difficult for the model to classify.
[ ] Performance on New Images: the performance of the model when tested on the captured images.
[ ] Performance on New Images: The performance on the new images is compared to the accuracy results of the test set.
[ ] Model Certainty - Softmax Probabilities: The top five softmax probabilities of the predictions on the captured images are outputted.
[ ] Model Certainty - Softmax Probabilities: discusses how certain or uncertain the model is of its predictions.
Image processing for training the network. Note the many random
distortions applied to the image.
Randomly crop a [height, width] section of the image. distorted_image = tf.random_crop(reshaped_image, [height, width, 3])
Randomly flip the image horizontally. distorted_image = tf.image.random_flip_left_right(distorted_image)
Because these operations are not commutative, consider randomizing
the order their operation. distorted_image = tf.image.random_brightness(distorted_image, max_delta=63) distorted_image = tf.image.random_contrast(distorted_image, lower=0.2, upper=1.8)
Subtract off the mean and divide by the variance of the pixels. float_image = tf.image.per_image_whitening(distorted_image)